CULTIVAR AND GERMPLASM RELEASES

HORTSCIENCE 52(4):649–651. 2017. doi: 10.21273/HORTSCI11692-16 on six representative canes collected annu- ally in March, each with eight nodes to determine primary bud survival. Buds were ‘Itasca’ Grapevine, a New Cold-hardy sliced with a razor blade and the meristem evaluated for living tissue (green) or necro- Hybrid for White Wine Production sis using a dissecting microscope. The pri- 1 mary bud survival in Spring 2014 for ‘Itasca’ Matthew Clark , Peter Hemstad, and James Luby was 64.6%. This compares to ‘Frontenac Department of Horticultural Science, University of Minnesota, St. Paul, gris’ with 25.0%, ‘’ 39.6%, and MN 55108 ‘’ with 14.6% primary bud sur- vival. The 2013–14 winter weather season Additional index words. cultivar, fruit breeding, , Vitis was severe in Minnesota, with the lowest temperature reported on 6 Jan. 2014 at –31.1 C. Weather station data were collected from the ‘Itasca’ is a new grape cultivar for wine- with superior winterhardiness (Fig. 2). In cli-MATE database (mrcc.isws.illinois.edu/ making, which has been bred for use in cold addition to V. riparia, the pedigree of ‘Itasca’ CLIMATE) for Chanhassen WSFO, MN climates due to its ability to survive and contains the following Vitis species: V. aestivalis, 5 km from the vineyard area. In addition to produce high-quality fruit in USDA hardi- V. berlandieri, V. cinerea, V. labrusca, the extreme low temperatures, there were ness zone 4 (average minimum –37.2 to V. rupestris,andV. vinifera. The seedling periods of extended cold that may have –34.4 C). The vines have demonstrated im- vine was identified in 2009 for exhibiting resulted in greater bud mortality in ‘Frontenac proved winterhardiness compared with other improved fruit quality, cold-hardiness, and gris’, ‘La Crescent’, and ‘Marquette’. The cold climate cultivarssuchasMarquette,La disease resistance attributes and was desig- primary bud survival in Spring 2015 was as Crescent, and the Frontenac series. ‘Itasca’ nated as selection MN 1285 for testing, and follows: ‘Itasca’ 95.8%, ‘Frontenac gris’ exhibits disease and insect resistance, which propagated by hardwood cuttings. 81.3%, ‘La Crescent’ 75.0%, and ‘Marquette’ may make it a candidate for reduced spray 43.8%. The lowest temperature reported in programs using integrated pest management Description and Performance Winter 2014–15 was on 13 Jan. 2015 at or organic approaches. At harvest, Itasca fruit –27.2 C. The primary bud survival in Spring has improved juice chemistry compared with Growth and yield. ‘Itasca’ was compared 2016 was as follows: ‘Itasca’ 91.6%, ‘Fronte- the previously listed cultivars, making it with other cold-hardy, hybrid wine nac gris’ 83.3%, ‘La Crescent’ 95.8%, and more suitable for wine production specifi- including its parent ‘Frontenac gris’ for the ‘Marquette’ 89.6%. The lowest temperature cally a reduction in total titratable acidity growth and yield observations. The research reported in Winter 2015–16 was on 17 Jan. (TTA) compared with other cold-hardy, vineyards have a minimal spray program to 2016 at –27.2 C. white wine cultivars Frontenac gris and La allow for evaluation of resistance to pests and Disease resistance. ‘Itasca’ vines were Crescent. ‘Itasca’ has more sugar at harvest diseases. Vines were grown on their own grown in vineyards with an integrated pest than ‘Brianna’, another cold-hardy variety roots and evaluated as top wire, high-cordon management approach. Preventative fungi- with low total acidity levels. vines grown bilaterally with multiple trunks cide sprays were limited to no more than as is common practice in Minnesota. Vines three applications per season. Insecticides Origin were dormant pruned, and weeds controlled were applied only after scouting revealed through mechanical and chemical methods significant damage from pests such as grape The grapevine ‘Itasca’ (Fig. 1) was iden- including pre- and postemergent herbicides. berry moth (Paralobesia viteana Clemens), tified at the Horticultural Research Station ‘Itasca’ has a more vigorous growth habit phylloxera (Daktulosphaira vitifoliae Fitch), (HRC), a Minnesota Agricultural Experi- than ‘Frontenac gris’, but less vigorous than or Japanese beetle (Popillia japonica ment Station located near Excelsior, MN # ## # ## ‘La Crescent’. This cultivar has forked ten- Newman). The vineyard plantings included (44 52 9.2496 N, –93 38 15.435 W). drils that are shorter in length than La highly susceptible cultivars and breeding The seedling vine originated from a cross Crescent tendrils. The trunk bark is moder- lines that were adjacent to these plants and made by P. Hemstad in 2002 (Fig. 2) between ately flaky with segments 5 mm · 5cmin serve as susceptible checks. The ‘Itasca’ the maternal plant ‘Frontenac gris’ (Vitis · length. vines were observed for disease and pest riparia 89 Landot 4511) (Luby and Hemstad, The average cluster weight for ‘Itasca’ is signs and symptoms. Over 6 years of evalu- 2003) and an advanced breeding selection, · 145 g, compared with ‘La Crescent’ which is ations (2011–16), the vines have shown a very MN 1234 (MN 1095 ‘’), as the 87 g, or ‘Frontenac gris’, which has average low incidence of powdery mildew (Uncinula pollen parent. Both parents are interspecific clusters weighing 124 g. ‘Itasca’ clusters necator), which is attributed to genetic re- hybrids derived from V. riparia accessions have an average of 86 berries per cluster with sistance. Molecular markers have linked the mean weight of 1.27 g/berry. The average resistance at the Ren3 locus (Teh et al., 2017). yield per vine was calculated from harvest The Ren3 resistance has been overcome in Received for publication 23 Dec. 2016. Accepted data from 2010 to 2016 (Table 1). The high- some growing regions by local pathogen for publication 15 Feb. 2017. est yields of the observed varieties were in populations (Cadle-Davidson, personal com- This research was conducted in part with support of ‘Frontenac gris’ with 7.05 kg/vine, followed munication). No evidence of downy mildew the University of Minnesota Agricultural Experi- ment Station, the Viticulture Consortium East by ‘La Crescent’ with 6.17 kg/vine, and then (Plasmopara viticola)orBotrytis has been (USDA 2009-34360-19879), VitisGen (USDA/ ‘Itasca’ with 5.29 kg/vine. The average yield observed. A slight amount of foliar black rot NIFA 2011-51181-30635), and Northern Grapes for ‘Marquette’ was lower with 4.95 kg/vine (Guignardia bidwellii) has been seen in some Project (USDA/NIFA 2011-51181-30850). on high cordon and 2.70 kg/vine on vertical years. Limited foliar infestation by the insect We thank the expert vineyard team Jennifer shoot positioned vines (VSP) (Table 1). pest phylloxera was observed in 2016. Less Thull and John Thull for their technical support ‘Brianna’ had the lowest yield per vine at than five leaves were identified with phyllox- in vine management and propagation. Wine- 3.54 kg. In Minnesota, it can be as little as 1 era galls, compared with ‘Frontenac’, which making was conducted by Nicholas Smith, week between budburst between ‘Marquette’ is highly susceptible and had many shoots Kathryn Cook, Drew Horton, and Bryan Forbes. and ‘Frontenac gris’, with ‘Itasca’ budburst entirely infested with galls. A single 2-year- Genotypic information about ‘Itasca’ was ac- quired through the VitisGen project (USDA/ between the two. The average bloom time in old vine-grown VSP trellis system exhibited NIFA 2011-51181-30635) and conducted by Excelsior, MN, was 15 June, about the same trunk injury, although no indication of crown Soon Li Teh. as ‘Marquette’. gall infestation (Agrobacterium vitis) was 1Corresponding author. E-mail: clark776@umn. Winterhardiness. A destructive sampling detected. In the very wet growing season of edu. technique was used to dissect compound buds 2016, bunch rots (of undiagnosed type) were

HORTSCIENCE VOL. 52(4) APRIL 2017 649 observed in Itasca as well as many other should be at least 1.83 m between plants The wine was settled for 2–5 d before cultivars. and it is recommended that growers evaluate racking, and adjusted to 40 ppm FSO2 and Canopy management. The moderately their own site conditions. Pruning weight cold stabilized at 1 C for 14 d before vigorous growth of ‘Itasca’ on the high wire data were not collected due to vineyard bottling. cordon system in fertile soils (high organic practices such as shoot thinning, which Juice and wine analysis. Before harvest, matter) as tested at the HRC has shown that would confound results. Allowing 2-year- berries were monitored for maturity using a suitable planting distance in the row old vines to produce fruit in the establish- visual, organoleptic, and field chemistry ment years can help de-vigorate excessive analysis. All juice samples were measured vine growth, as observed in a VSP planting with a handheld digital refractometer established in 2015 at the HRC, without (Atago Pocket Pal-1; Atago USA, Belle- predisposing vines to winter injury. The vue, WA) to estimate soluble solids content ‘Itasca’ vine is amenable to VSP due to the (SSC; measured as Brix). The acid content presence of upright shoots along the cordons was estimated by measuring pH using the and the plasticity in training. However, no accumetä AR15 pH meter (Fisher Scien- data have been collected on fruit yield when tific, Hampton, NH), and by measuring grown as VSP. Dormant season pruning is TTA(expresseding·L–1 tartaric acid equiv- necessary to maintain vine yields and fruit alents) using the Mettler-Toledo DL28 quality through selective removal of fruiting titrator and the DG115-SC pH probe with spurs and canes. auto-sampling (Mettler-Toledo, Columbus, Enological performance. Wines have OH). Juice was analyzed for free amino been produced from ‘Itasca’ berries since nitrogen (FAN), yeast assimilable nitrogen 2009. A standard protocol was used for the (YAN), and malic acid concentration by microvinification each year. Berries were enzymatic analysis (UniTABä, Unitech Sci- pressed as whole clusters in a manual, entific, HIan Gardens, CA). A trained sensory basket-style screw press or bladder hydro panel provided ratings and descriptors for the press depending on the harvest volume in wines. each year. The pressed juice was settled for Juice data at harvest. At harvest, juice 24 h at 13 C, then racked and allowed to samples were evaluated for sugar and acid come to 20 C. The juice was amended with content (Table 1). Over the six vintages from yeast nutrients Go-Fermä (0.5 g·L–1; Danstar 2010 to 2015, ‘Itasca’ was harvested between Ferment AG, Zug, Switzerland) and Fermaid 12 and 28 Sept., with a typical harvest around Kä (0.15 g·L–1; Danstar Ferment AG), and 24 Sept. (Table 1). At the HRC, ‘Itasca’ is inoculated with the yeast Lalvin DV10 (0.4 picked earlier than ‘Frontenac gris’. The g/L; Danstar Ferment AG, Fredericia, Den- mean SSC of ‘Itasca’ is similar to ‘Frontenac Fig. 1. Photograph of ‘Itasca’ grapevine fruit mark) supplied by Scott Laboratories, Inc. gris’, but higher than ‘La Crescent’ and cluster at maturity (photo by David Hansen, Petaluma, CA. Fermentation was conducted ‘Brianna’ (Table 1). The range for SSC for University of Minnesota Agricultural Experi- for 24 h at 20 C before moving to 13 C ‘Itasca’ at harvest ranged from 24.8 Brix in ment Station). to complete the fermentation (to dryness). 2014 to 28.2 Brix in 2012. The TTA for ‘Itasca’ was lowest in 2012 with 8.32 g·L–1 but was highest in 2014 at 11.53 g·L–1. The lower mean TTA and lower malate values, in combination with high SSC, set this cultivar apart from other cold-climate cultivars. The means for juice malate concentrations from fruit harvested for each cultivar were as follows: Itasca (3.05 g·L–1), Frontenac gris (6.43 g·L–1), La Crescent (7.33 g·L–1), Brianna (2.8 g·L–1), and Marquette (4.63 g·L–1) (Table 1). The lower concentration of malic acid in ‘Itasca’ juice is favorable for wine making. ‘Brianna’, another low-acid wine had much lower sugar (5.5 Brix less) at harvest than ‘Itasca’ (Table 1). Nitrogen Fig. 2. Pedigree of ‘Itasca’ grapevine, truncated to demonstrate three generations, including two distinct Vitis content of juice and must is highly variable riparia ancestors (V. riparia MN 89 = PI588262). Line color indicates parental configuration (red = across cultivars and locations (Butzke, 1998; female, blue = pollen). Hagen et al., 2008), but adequate nitrogen is

Table 1. Mean values and SD for yield per vine and juice chemistry attributes at harvest for vintages 2010–15.z Yieldz TTA Malate SSC FAN YAN Cultivar (kg) SD pH SD (g·L–1) SD (g·L–1) SD (Brix) SD (mg·L–1) SD (mg·L–1) SD Harvest window Itasca 5.29 0.98 3.20 a 0.13 9.54 c 1.13 3.05 c 1.63 26.2 a 1.26 89 c 32.7 94 c 31.6 9/12–9/28 Frontenac gris 7.05 1.89 3.09 ab 0.08 13.05 ab 0.54 6.43 ab 1.33 26.0 a 1.65 215 a 39.4 225 a 36.9 9/11–10/14 La Crescent 6.17 2.21 3.05 ab 0.08 14.45 a 1.46 7.33 a 1.10 22.7 b 1.46 131 bc 3.5 135 bc 1.4 9/14–10/2 Brianna 3.54 1.35 3.16 ab 0.21 7.74 d 1.48 2.8 c 1.22 20.7 c 0.97 162 ab 33.2 166 ab 37.0 8/30–10/6 Marquette 4.95 2.96 3.04 b 0.11 12.04 b 1.30 4.63 bc 0.79 25.3 a 1.69 159 ab 29.7 178 ab 38.7 9/6–9/29 Measures of acidity [pH, total titratable acidity (TTA) as tartaric acid equivalents], soluble solid content (SSC), free amino nitrogen (FAN), yeast assimilable nitrogen (YAN), and harvest range over the vintages. Cultivar means within columns with different letters are significantly different (P < 0.05) with respect to Fisher’s least significant difference (LSD) post hoc analyses. z Yield per vine is estimated from mature vines on 6- or 8-foot spacing in several vineyard locations across years 2010–16, where available. No Fisher’s LSD provided due to unbalanced design.

650 HORTSCIENCE VOL. 52(4) APRIL 2017 Table 2. Mean values and SD for wine chemistry attributes after fermentation to dryness using a standard Availability winemaking protocol for vintages 2010–13.

Cultivar pH SD TTA (g·L–1) SD % ethanol (v/v) SD The University of Minnesota has filed an Itasca 3.17 b 0.04 7.62 b 1.56 14.4 a 0.82 application for a U.S. Plant Patent on the Frontenac gris 3.28 ab 0.31 11.09 a 1.12 14.1 a 1.11 ‘Itasca’ grapevine. Nurseries wishing to grow La Crescent 3.32 ab 0.10 11.06 a 0.27 13.1 a 0.27 and sell ‘Itasca’ plants must first sign a license Brianna 3.32 ab 0.19 7.23 b 1.01 13.5 a 1.01 agreement, which can be arranged with the Marquette 3.50 a 0.05 6.78 b 0.64 14.2 a 0.64 University of Minnesota Office of Technol- Cultivar means within columns with different letters are significantly different (P < 0.05) with respect to ogy Commercialization. More information Fisher’s least significant difference post hoc analyses. can be found at http://mnhardy.umn.edu/ commercial-growers/license/license-grapes. Contact B.J. Haun ([email protected]) critical in winemaking for yeast nutrition to Juice and wine sensory attributes. The for a list of current licensees or to inquire prevent stuck fermentation. The relatively ‘Itasca’ berries are pleasant for fresh eat- about a license agreement for the propaga- low FAN and YAN values for ‘Itasca’ ing, but contain fully developed seeds and tion and sale of ‘Itasca’ grape vines. (Table 1) provide evidence that yeast nutri- have a high seed to flesh ratio. The berries tion should be monitored and supplemented are also much smaller than conven- Literature Cited throughout winemaking where needed (Bely tional table grapes such as ‘Swenson et al., 1990). Red’. The fruit flavors have been described Bely, M.B., J.-M. Sablayrolles, and P. Barre. 1990. Wine chemistry data. Following fermen- as subtle notes of pear, gooseberry, honey- Automatic detection of assimilable nitrogen tation, the wines were measured for pH, TTA, dew melon, and star fruit. The flavor attri- deficiencies during alcoholic fermentation in and % ethanol (Table 2). After fermentation, butes of the juice translate to the wine, oenological conditions. J. Fermentation and Biogengineering. 70(4):246–252. the pH of the wines made from ‘Itasca’ was although the wine has been described as Butzke, C.E. 1998. Survey of yeast assimilable nitro- within a normal and acceptable range. somewhat neutral. Sensory analysis has gen status in musts from California,Oregon, and Through fermentation, the TTA of the wines elicited wine descriptors such as pear, Washington. Amer. J. Enol. Viticult. 49:220–224. exhibited an average reduction of 1.92 g·L–1. quince, star fruit, melon, and subtle honey Hagen, K.M., M. Keller, and C.G. Edwards. 2008. The major reduction in TTA in ‘Marquette’ notes. ‘Itasca’ differs from other hybrid Survey of biotin, pantothenic acid, and assimi- can be attributed to the malolactic fermen- varieties and does not have the typical flavor lable nitrogen in wine grapes from the Pacific tation (secondary fermentation) that is im- profile of V. labrusca wines like ‘Brianna’ or Northwest. Amer. J. Enol. Viticult. 59:432–436. plemented for red wines but was not ‘Edelweiss’. The wine color is golden to Luby, J. and P. Hemstad. 2003. Grape plant named implemented for the other cultivars. The light green. Because of the light and subtle ‘Frontenac gris’. USPP16478. Teh, S.L., J. Fresnedo-Ramírez, M.D. Clark, D.M. percent ethanol by volume of the finished flavors, as well as reduced acidity compared Gadoury, Q. Sun, L. Cadle-Davidson, and J. ‘Itasca’ wine was 14.4 on average. The with ‘Frontenac gris’ or ‘La Crescent’, Luby. 2017. Genetic dissection of powdery amount of ethanol varied depending on the ‘Itasca’ is likely to be a high-quality varietal mildew resistance in interspecific half-sib season, and was directly related to the SSC of wine or used in wine blends to improve grapevine families using SNP-based maps. the juice at harvest. mouth feel and flavor. Mol. Breeding. 37:1.

HORTSCIENCE VOL. 52(4) APRIL 2017 651